DeepCoder-14B competes with frontier models like o3 and o1—and the weights, code, and optimization platform are open...
AI research
Auto Added by WPeMatico
Reward models holding back AI? DeepSeek’s SPCT creates self-guiding critiques, promising more scalable intelligence for enterprise LLMs.Read...
CoTools uses hidden states and in-context learning to enable LLMs to use more than 1,000 tools very...
Anthropic has developed a new method for peering inside large language models like Claude, revealing for the...
METASCALE uses a three-stage approach to dynamically choose the right reasoning technique for each promblem.Read More
With multiple sampling and self-verification, Gemini 1.5 Pro can outperform o1-preview in reasoning tasks.Read More
SEARCH-R1 trains LLMs to gradually think and conduct online search as they generate answers for reasoning problems.Read...
A new framework inspired by the RICE scoring model balances business value, time-to-market, scalability and risk for...
Chain-of-experts chains LLM experts in a sequence, outperforming mixture-of-experts (MoE) with lower memory and compute costs.Read More
A-MEM uses embeddings and LLMs to create dynamic memory notes that automatically link to create complex knowledge...
